Proc. NatL Acad. Sci. USAVol. 79, pp. 4756-4760, August 1982Immunology

Mediators from cloned T helper cell lines affect immunoglobulinexpression by B cells

(B cell lines/T cell supernatants/immunoglobulin synthesis and secretion)

CHRISTOPHER J. PAIGE, MAx H. SCHREIER, AND CHARLES L. SIDMANBasel Institute for Immunology, Grenzacherstrasse 487, CH-4005 Basel, Switzerland

Communicated by N. K. Jerne, May 4, 1982

ABSTRACT When cloned T helper cells encounter antigenpresented by I-A-compatible macrophages, soluble mediators areproduced that affect the differentiation and activation of normalB lymphocytes and cell lines of the B lineage. Exposure to suchT cell culture supernatants causes two effects in the murine70Z/3 cell line, which represents apre-B stage of differentiation.These cells begin to synthesize Ig light chains and gain membraneIg that is detectable by immunofluorescence. Two other effectsare seen after similar treatment of the WEHI-279. 1 murine cellline, which represents a mature, Ig+ B cell. These cells shift theratio of ,l chains produced from mostly membrane to mostly se-cretory type and begin to secrete large amounts ofIgM, which canbe detected either by biosynthetic radiolabeling followed by im-munoprecipitation or by a staphylococcal protein A plaque assay.The majority also die. Similar to WEHI-279.1, normal small rest-ing B cells also show the shift from membrane ,u to secretory ,uand are activated to Ig secretion after exposure to these super-natants. These results show that products from T cell immune re-actions exert multiple effects on B cell development and activation,at several stages of the B cell developmental pathway. The ob-served changes range fromnuclear processes, including gene tran-scription and RNA splicing, to such post-translational aspects asprotein processing, catabolism, membrane architecture, and cellsurvival.

Supernatants of stimulated lymphoid cell cultures contain bi-ologically active factors that influence the growth and differ-entiation of B lymphocytes (reviewed in refs. 1 and 2). In thepast, it has been difficult to precisely define the biological func-tion of these activities and determine at which level of B celldevelopment they operate. In part this is due to the complexcellular systems that have been used (3). Some of these diffi-culties have been overcome by the use of supernatants ofcloned, antigen-specific T cell lines, cultured in the presenceof appropriate antigen and histocompatible accessory cells (4,5).The first recognizable cells of.the B lineage are cytoplasmic

Ig+ cells that synthesize u.heavy (H) but not light (L) chains andare found in fetal liver (6, 7). The asynchronous onset in Ig chainsynthesis is reflected at the level of DNA, because the geneticrearrangements necessary for Ig expression occur first for the,u gene and subsequently for the L chain gene (8, 9). It is be-lieved that these ",u-only" cells give rise to progeny with fullyrearranged L chain genes, although L chain may not yet be syn-thesized (8, 10). At this stage p. chains of the membrane formmay in fact be present on the cell surface in the absence of Lchain (10, 11). The following stage of differentiation is heraldedby the onset ofL chain synthesis and the expression ofcompleteIgM molecules on the cell membrane. These surface-Ig+ (sIg')cells are small and quiescent, in contrast to earlier stages, which

are composed oflarger proliferating cell types (12). At this pointin B cell development, the entire membrane A biosyntheticpathway is operational, while only the early stages of the se-cretory p. pathway are present (13). Subsequent to activation,additional post-translational processes generate the missing se-cretoryp. forms and allow IgM secretion (13). Finally, in themost mature cells of the B lineage, Ig-secreting plasma cells,the membrane pa pathway may be reduced or deleted (13-15).

In the present report, we demonstrate that factors containedin the supernatants of antigen-stimulated cloned T cells causedifferentiation ofestablished cell lines ofthe B lineage. The pre-B cell line 70Z/3 (10, 16), which under normal growth condi-tions synthesizes immunoglobulin pu H chains but not L chains,is induced to synthesize L chains, which are subsequently ex-pressed on the cell surface. The B cell lymphoma WEHI-279. 1(17, 18), which under normal growth conditions synthesizes anddisplays membrane IgM while secreting only low amounts ofIgM extracellularly, is induced to secrete large amounts ofIgM.Studies with both small resting splenic B cells and WEHI-279. 1also demonstrate a rise in the secretory-to-membrane p. ratioand increased cell death after exposure to T cell supernatants.These assays provide easy and reproducible means to screensmall amounts of biologically.active material and also allow thedetailed analysis of the effects of soluble factors on multiple.stages of B cell differentiation.

MATERIAL AND METHODSB Cell Lines. The murine pre-B cell line 70Z/3 was estab-

lished in vitro from the spleen of a thymectomized (C57BL/6x DBA/2)F1 mouse previously exposed to methylnitrosourea(16, 19). The growth properties of this line and its response tomitogen have been described (10, 16). Murine. WEHI-279.1cells were a generous gift from W. Raschke (La Jolla, CA) andhave also been previously described (17, 18). Additional murineB cell lines tested in these experiments include WEHI-231 (20),GCL-2.8 (21), and 38C-13 (22). These lines were routinely cul-tured in Iscove's modified Dulbecco's medium (IMD medium)supplemented with human transferrin at 5 Amg/ml, 0.5% delip-idated bovine serum albumin, and 5% fetal calf serum (Boeh-ringer Mannheim no. 66459502).T Helper Cell Lines. Helper T cell lines were established

by repeated antigen stimulation in vitro in the presence of ir-radiated spleen cells, cloned under conditions of limiting di-lution, and propagated in T cell growth factor (TCGF)-supple-mented medium as described (23, 24). The supernatants usedin this study were derived from 60-hr serum-free cultures con-

Abbreviations: L and H chains, light and heavy chains of immunoglob-ulins; y, and P, secretory and membrane a H chains; slg, surface im-munoglobulin; IMD medium, Iscove's modified Dulbecco's medium;SRBC, sheep erythrocytes; OVA, ovalbumin; PE, peritoneal exudatecells; LPS, lipopolysaccharide; PFC, plaque-forming cells; kDal,kilodalton(s).

4756

The publication costs ofthis article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertise-ment" in accordance with 18 U. S. C. §1734 solely to indicate this fact.

Dow

nloa

ded

by g

uest

on

Aug

ust 1

5, 2

021

Proc. Natl. Acad. Sci. USA 79 (1982) 4757

taining (per ml) 2.5 x 104 sheep erythrocyte (SRBC)-specifichelper T cells (C57BL/6) or 5 x 104 chicken ovalbumin (OVA)-specific helper T cells (C57BL/6), the antigen SRBC (2.5 x 106)or OVA (100 /ug), and 1 X 104 C57BL/6 nu/nu spleen cells or

5 X 10' peritoneal exudate cells (PE) derived from eitherC57BL/6 normal (+/+) or nude (nu/nu) mice or the indicatedcongeneic mouse strains.

Cells and Cultures. In most experiments 1-5 X 105 WEHI-279.1 cells or 5 X 10' 70Z/3 cells were grown in 0.5 ml of me-dium in 24-well tissue culture plates (Costar, Cambridge, MA).Small spleen cells were isolated from erythrocyte-depleted pop-ulations at .1.08 g/cm3 in Percoll density gradients and cul-tured at a concentration of 106 per ml. To each well was addedeither T cell supernatant or a similar volume of medium. Thecells were incubated for 48 hr at 37TC in 5% CO2. After culturethe cells were harvested and washed once in Earle's balancedsalts solution, and viable cells were counted on the basis of try-pan blue exclusion. Aliquots were used for the various assays.

Similar cultures were exposed to lipopolysaccharide (LPS, fromSalmonella typhosa W0901, Difco) at 10-100 /Lg/ml for celllines and 50 tug/ml for small spleen cells. 70Z/3 cells and smallspleen cells consistently responded to LPS under these con-

ditions. The response of WEHI-279.1 cells to LPS was more

variable and was found to be dependent on undefined factorspresent in different batches of fetal calf serum (unpublishedresults). All batches of fetal calf serum tested were able to sup-

port LPS responses ofspleen cells and 70Z/3 cells. Small spleencells, 70Z/3 cells, and WEHI-279.1 cells consistently re-

sponded to T cell supernatants in all batches of fetal calf serumtested and serum-free conditions.

Immunofluorescence. The sIg expression of70Z/3 cells wasdetermined by using a rhodamine-conjugated anti-A reagentprepared from a rabbit anti-MOPC 104E antiserum and en-

riched by binding to staphylococcal protein-A-conjugated Seph-arose (donated by L. Forni, Basel). Generally, 5 X 105 cellswere washed once and resuspended in 50 A.l of staining reagentfor 15 min on ice. They were subsequently washed once withIMD medium, once through 100% fetal calf serum, and a thirdtime with IMD medium containing 2% fetal calf serum. Thecells were then resuspended in one or two drops of IMD me-

dium containing 2% fetal calf serum and examined with a ZeissPhotomicroscope.

Plaque Formation. Protein A-coupled SRBC were preparedaccording to the method of Gronowicz et al (25). The reverse

plaque assay, using appropriately diluted rabbit anti-mouse IgMantiserum (anti-MOPC 104E) and selected guinea pig comple-ment (GIBCO), was used as described (25).

Biochemical Analysis. After culture, cells were washed twicein Earle's balanced salts solution. Equal numbers of cells werelabeled with [3H]leucine or [35S]methionine, in medium defi-cient in that amino acid, for 1 hr in the absence of tunicamycin(for the patterns of glycosylated molecules in cell lysates), for4 hr in the absence of tunicamycin (for secreted glycosylatedmolecules in culture supernatants), or for 3 hr in the presence

of tunicamycin at 5 pug/ml (after 1-hr preexposure to tunica-mycin) (for the pattern of nonglycosylated molecules in cell ly-sates). After these labeling periods, culture supernatants or

nonionic detergent cell lysates were immunoprecipitated withrabbit anti-mouse IgM (A and K) antiserum and protein A-Seph-arose CL4B beads, eluted, electrophoresed on 20-cm Na-DodSO4/linear 8-18% polyacrylamide gradient gels, and fluo-rographed. Some gel bands were excised and their radioactivitieswere measured. These procedures may be found in detail in ref.13.

RESULTSExposure of the pre-B cell line 70Z/3 to supernatants derivedfrom T helper cells stimulated by antigen in the presence of I-A-compatible accessory cells resulted in a significant increasein sIg expression as judged by immunofluorescence. Data ob-tained with two helper T cell clones (24), OVA-7, specific forOVA, and S26-5, specific for SRBC, are shown in Table 1. Thesedata indicate that the induction ofactive T cell supernatants wasantigen specific and required I-A-compatible accessory cells.We also investigated the differentiation of B lymphocytes in

response to T cell supernatants by biosynthetically labeling Bcells with radioactive amino acids and examining the resultingintracellular and secreted molecules by immunoprecipitationand NaDodSO4/polyacrylamide gel electrophoresis. Resultsobtained with small splenic B cells and with cell lines 70Z/3 andWEHI-279. 1 are shown in Fig. 1.

The first radiolabeling protocol was a 1-hr incubation in me-dium containing radioactive amino acids followed by examina-tion of the intracellular Ig molecules. Previous studies haveshown that no detectable processing or secretion ofradiolabeledIg molecules occurs in this period and that the resulting pat-terns, especially of A chains, are diagnostic of the B cells' stateof differentiation (13). As expected, small (resting) splenic Bcells produced one major glycosylated intracellular ,u species,of apparent molecular mass 73 kilodaltons (kDal) (Fig. 1, 1-hrlysate). After exposure to LPS or T cell supernatants, the twoglycosylated intracellular a species (73 and 70 kDal) character-istic of activated B cells were seen instead. It has been previ-ously demonstrated that the 70-kDal form is unique to the se-cretory pathway, whereas both secretory and membrane ,u have73-kDal forms (13). T cell supernatants, but not LPS, also in-creased the 70-kDal form in WEHI-279. 1 cells. Neither T cellsupernatant nor LPS detectably altered ,u chain biosynthesisin 70Z/3 cells, although both induced the synthesis of L chains(25 kDal). (IThe induction of L chains in 70Z/3 is better seen inthe 3-hr radiolabeling results with tunicamycin-see below.)The induction of the activity that causes L chain synthesis alsorequired antigen-mediated, I-A-restricted T cell-macrophageinteraction (Table 1).

Table 1. Effects of T cell supernatants on sIg and L chainexpression in 70Z/3 cells

Presenceof L

T cells Antigen Accessory cells % sIg+ chain

OVA-7 OVA B6 +/+ PE 81 +OVA-7 OVA B6 nu/nu PE 88 +OVA-7 OVA None 12OVA-7 OVA B10.A (4R) PE* 21 -OVA-7 OVA B10.A (5R) PE 83 +OVA-7 OVA B10.MBR PE* 13 -OVA-7 None B6 nu/nu PE 18OVA-7 SRBC B6 nu/nu spleen 16S26-5 SRBC B6 nu/nu spleen 82 +

70Z/3 cells (5 x 10' per ml) were cultured for 48 hr in IMD mediumcontaining 10% fetal calf serum and 20% supernatants derived from60-hr cultures containing, per ml, 2.5 x 104 SRBC-specific helper Tcells (S26-5) or 5 x 104 OVA-specific helper T cells (OVA-7), the an-tigen [SRBC (2.5 x 106) or OVA (100 ,ug)], and 5 x 104 PE of the in-dicated congeneic mouse strains or 1 x 104 C57BL/6 (B6) nu/nu spleencells. sIg expression was determined by using rhodamine-conjugatedrabbit anti-mouse ,u antibody. The presence of L chain was determinedby following the protocol described for Fig. 1. Uninduced control cul-tures were 10% sIg' and L chain negative.* These strains are I-region nonidentical with the C57BL/6 helper Tcell clones.

Immunology: Paige et aL

Dow

nloa

ded

by g

uest

on

Aug

ust 1

5, 2

021

Proc. Natl. Acad. Sci. USA 79 (1982)

1-hr lysate 3Sm 279 170Z-LT --LT-LT

7370-

6260

25 -2

3-hr lysate ±+Tm)Sm 279170Z

-- LT-LT-LT

4-hr sup.Sm 12791 70Z-LT-LT- LT

FIG. 1. Biochemical analysis of Ig production. Small spleen cells(Sm), WEHI-279.1 (279), and 70Z/3 (70Z) were cultured for 48 hr inthe presence of LPS (L), 20% T cell supernatant (T), or without addedstimulus (-). The first panel (1-hr lysate) shows the electrophoreticpatterns of the intracellular glycosylated molecules. The second panel[3-hr lysate (+Tm)] shows the intracellular nonglycosylated moleculeslabeled for 3 hr in the presence of tunicamycin (5 ,ug/ml) (after 1 hrpreexposure to tunicamycin). The third panel (4-hr sup.) shows thepattern of secreted glycosylated molecules in the culture supernatants.Molecular masses in kDal are indicated.

Because the synthesis of the intracellular 70-kDal A formcorrelates with and is probably necessary for IgM secretion (13),we examined culture supernatants for secreted IgM (Fig. 1, 4-hr sup.). Both LPS and T cell-derived supernatants greatly in-creased the IgM secretion of small splenic B cells. (In the ex-periment shown, 2,900, 28,100, and 17,000 cpm ofa chain weresecreted by control, LPS- and T cell supernatant-treated cells,respectively.) Cell line WEHI-279. 1 responded with increasedA secretion to T cell supernatants but not to LPS (2,100, 1,100,and 30,900 cpm of u chain, respectively). Cell line 70Z/3 nevershowed appreciable A secretion (600, 400, and 300 cpm of A,respectively).We also measured the relative amounts of membrane and

secretory A (Am and ,s, respectively) produced by examiningthe nonglycosylated polypeptides made in the presence of tuni-camycin. Both forms (62 and 60 kDal) are present in resting Bcells, but the ratio changes in favor of the secretory form (60kDal) after B cell activation (13), probably reflecting changesin the corresponding mRNA populations (14, 15). The ratio ofsecretory to membrane , made in small B cells changed in re-sponse to both LPS and T cell supernatants [1.4, 3.0, and 3.0,S per gm in small B cells treated with medium alone, LPS, andT cell supernatants, respectively, in the experiment shown inFig. 1, 3-hr lysate (+Tm)]. In WEHI-279.1 populations, thesecretory-to-membrane ,u ratio changed only in response toT-cell supernatants and not to LPS (0.67, 0.72, and 2.9 /is perAm, respectively). With 70Z/3 populations, neither treatmentchanged the secretory-to-membrane ,u ratio (0.45, 0.41, and0.52 ,s per Am, respectively).The effect of T cell supernatants on WEHI-279. 1 cells was

also demonstrated in a protein A reverse plaque assay (Table2). Both plaque formation and the shift in the proportions of ,uforms required T cell supernatants generated with I-A-com-patible accessory cells and appropriate antigen. We also ob-served that the recovery of WEHI-279. 1 cells was greatly di-minished in the cultures containing active T cell supernatants,

Table 2. Effects of T cell supernatants on the growth and Igexpression of WEHI-279.1 cells

PFC/106 Cellsviable recovered

Tcells Antigen Accessory cells cells p/pA x 106OVA-7 OVA B6 +/+ PE 900 2.2 1.1OVA-7 OVA B6 nu/nu PE 1,100 2.8 1.2OVA-7 OVA None 3 0.56 13OVA-7 OVA B10.A (4R) PE* 150 1.0 4.6OVA-7 OVA B10.A (5R) PE 1,200 4.0 2.3OVA-7 OVA B10.MBR PE* 100 1.5 8.1OVA-7 None B6 nu/nu PE 8 0.75 11OVA-7 SRBC B6 nu/nu spleen 30 0.96 9.5826-5 SRBC B6 nu/nu spleen 2,200 2.4 1.6

WEHI-279.1 cells (5 x 104 per ml; 6.25 x 105 total cells) were cul-tured for 48 hr in IMD medium containing 10% fetal calf serum and25% supernatants, as described forTable 1. Plaque-forming cells (PFC)were determined in a reverse staphylococcal proteinA assay (25). Theratio of p to A was determined. Uninduced control WEHI-279.1 cellsgenerated 6 PFC/106 viable cells and p/Am = 0.96.* These strains areI-region nonidentical with theC57BL/6 (B6) helperT cell clones.

as compared to noninduced control cultures. This effect is dueto extensive cell death rather than lack ofproliferation, becausea large number of nonviable cells were detected in thesecultures.

Fig. 2 details several aspects of the response of WEHI-279. 1cells to S26-5 supernatants. Fig. 2C shows that 1% supernatantinduced significant plaque formation per culture, and that pla-teau values were reached by 7.5% supernatant. The minimalcell recovery, shown in Fig. 2B, occurred at 15% supernatant.

104

a)

UC.)

.0-0 103

V-l

Z441020

A

0 1 10 100

a)

-

ci

w 104a)a)

10304)

13103 C

0 1 10 100

1

0

,B

0 1 10 100

D

0 1 10 100% supernatant

FIG. 2. Response of WEHI-279.1 cells incubated for 48 hr with in-creasing amounts of S26-5 T cell supernatant. Mean values + SEMare presented. (A) Number of PFC/106 viable cells. (B) Total numberof viable cells recovered per culture. (C) Total number of PFC/cul-ture. (D) ,.m Total ,u cpm were 3,200 ± 270 for 0% supernatantand 190 ± 10 for 9% supernatant.

4758 Immunology: Paige et aL

ICt,

Dow

nloa

ded

by g

uest

on

Aug

ust 1

5, 2

021

Proc. Natl. Acad. Sci. USA 79 (1982) 4759

100

50

0 -

0 0.01 0.1 1 10% supernatant

When these data are combined to yield the number of plaquesper 106 viable cells (Fig. 2A), maximal levels were reached at15%, reflecting the cell recovery pattern. The increase in theIJs/m ratio is shown in Fig. 2D; it continued to rise with in-creasing amounts of supernatant.

The response of 70Z/3 cells to increasing concentrations ofS26-5 supernatants is depicted in Fig. 3. This shows that 1%supernatant was sufficient to reach plateau levels for both sIgexpression (Fig. 3A) and L chain synthesis (Fig. 3B), and that

A0

_a 0

100

50

, =M "1 2 3 4

C

2.5

103 I0

-\ 1.5

1020 0

0 _ .5 r

1 2 3 4 1Time in culture, days

B

0

0

1 2 3 4

D

0

_~~~~

so

3 4

2 34

FIG. 4. Kinetics of responses to S26-5 supernatant. The superna-tant concentrations used with WEHI-279.1 and 70Z/3 cells were 15%and 9%, respectively. Data points represent individual experimentvalues. (A) Percent sIg+ 70Z/3 cells. (B) Amount of L chain synthesisby 70Z/3 cells in relation to total ,u synthesis. During this time period,u synthesis by 70Z/3 cell cultures remained constant within a factorof 2. Average ,u chain radioactivity per induced culture on days 1-4was 2,060, 1,700, 1,200, and 880 cpm, respectively, and the comparableL chain values were 1,200, 1,600, 1,400, and 1,000 cpm. Noninducedcultures produced 250 cpm or less of L chain. (C) Plaque formationby WEHI-279.1 cells, expressed as PFC/106 viable cells recovered. (D)pl,, in WEHI-279.1 cells.

B a0 '

0

*i00

0 0.01 0.1 -1410 0.01 0.1 1 10

FIG. 3. Response of 70Z/3 cells incubated for48 hr with increasing amounts of S26-5 super-natant. Data points are individual experimentalvalues. (A) Percent sIg' cells. (B) Amount of Lchain synthesis relative to total ja chain synthe-sis. Total A synthesis did not appreciably changein induced vs. noninduced cells: noninduced,2,500 ± 290 cpm; 9% supernatant, 1,700 ± 180cpm per culture. The comparable L chain valueswere 190 ± 60 for noninduced and 1,600 ± 180for 9% supernatant.

the activity in even supernatant diluted 1: 1,000 (0.1%) wasreadily detectable.We also analyzed the induction of 70Z/3 and WEHI-279. 1

cells over a 4-day time course (Fig. 4). The percentage of sIg+70Z/3 cells reached maximal levels by 24 hr (Fig. 4A) while Lchain synthesis (Fig. 4B), although substantial after 24 hr. re-quired 48 hr to reach maximum. Similarly, the change in theWEHI-279. 1 Ls//jm ratio was virtually complete by 24 hr (Fig.4D), but the PFC response required 48 hr to develop fully (Fig.4C).

DISCUSSIONPrevious studies have demonstrated that various biologicallyactive factors are found in supernatants of appropriately stim-ulated helper T cells. These activities can be measured in sev-eral in vitro systems including: (i) propagation of cloned T celllines (26, 27); (ii) replacement of T cells in antibody responsesto certain antigens (23, 24, 27); (iii) maintenance of antigen-in-dependent proliferation of B cell blasts (4); (iv) maturation ofsmall resting B cells to antibody secretion without proliferation(5); and (v) stimulation of granulocyte, macrophage, and eryth-rocyte progenitors to proliferate in semisolid medium (27-29).The observations reported here show that these supernatants

also contain molecules that affect multiple stages of Ig expres-sion by B cells and B cell lines. In the pre-B cell line 70Z/3,T cell supernatants, like LPS, cause L chain synthesis and sIgexpression that is detectable by immunofluorescence. The firsteffect is most likely due to a transcriptional change ofthe L chaingenes, because noninduced 70Z/3 cells contain very little Lchain mRNA (30, 31). The absolute amount of L chain synthe-sized per 70Z/3 culture rose at least 10-20 times in these ex-periments, while the amount of , produced stayed constantwithin a factor of2. The second change, the development of sIgdetectable by immunofluorescence, may merely represent analtered orientation of preexisting cell surface ,u chains, becausethese are detectable on noninduced 70Z/3 cells by surface io-dination and immunoprecipitation but not by immunofluores-cence (unpublished data). However, if these two changes aredirectly related, it may be of interest that the maximal numberof sIg+ cells was reached prior to maximal L chain synthesis.One should point out that the percentage of cells detected asIg+ is not a quantitative measure of the amount of surface IgMper cell. We do not yet know when maximal IgM levels per cellare reached.

At the following stage of B cell differentiation, that of theresting, sIg+ B lymphocyte, T cell supernatants yield two othereffects on Ig expression. One is a change in the proportion ofmembrane ,u synthesis to secretory ,u synthesis in favor of se-cretory ,u. This may be mediated by a perturbation in the RNAsplicing mechanism that produces the alternative mRNA spe-

A100

to

w 50-

0

100

btow

50

0

U,4104-)

'o0

0rM4

Immunology: Paige et aL

Dow

nloa

ded

by g

uest

on

Aug

ust 1

5, 2

021

Proc. Nati Acad. Sci. USA 79 (1982)

cies (14, 15). It should be noted that the total amount of A syn-thesized per viable WEHI-279. 1 cell did not increase signifi-cantly during this transition. The final effect on nonsecretingB cells is the completion of secretory , post-translational pro-cessing, as evidenced by the 70-kDal intracellular ,a species andactive IgM secretion (13). Maximal PFC per culture werereached with only a modest tu ratio change in the titrationexperiment, while maximal us/tim enhancement occurred be-fore significant PFC induction in the kinetic experiment. Theseobservations suggest that these two changes are distinct.The cell death caused by S26-5 supernatants is not yet under-

stood. All tested T cell supernatants caused this effect, and itwas seen over a wide range of initial cells per culture (5 X 104to 2 X 106). The high number of dead cells counted in the cul-tures indicates that proliferation did occur (usually 3- to 4-foldin 48 hr). The possibility that the induction of plaque formationcould be accounted for by selection and enrichment of the rarePFC in the uninduced population is ruled out by the very lownumber of background plaques (<10/106) when cultures wereinitiated, and the absolute increase in PFC/culture (Fig. 2C).These data may reflect a normal aspect of in vivo differentiationto Ig secretion and have therapeutic application as well. In ourhands, other B cell lines tested [WEHI-231 (20), GCL-2.8 (21),and 38C-13 (22)] were unresponsive, in all assays, to both LPSand T cell supernatants.

Other reports have shown changes in Ig expression by B celllines stimulated by polyclonal mitogens (16, 31-39) or super-natants from complex mitogen-induced cell cultures (40-42),but factors derived from antigen-induced helper T cell culturesor monoclonal T cell lines in general may be more physiologi-cally relevant and definable. At present, we do not knowwhether the many effects demonstrated here result from oneor several separate biologically active factors, or whether all ofthe observed changes within a given responding B cell popu-lation are aspects of a single integrated response pathway. Thesereadily measurable changes in stable B cell lines should, how-ever, provide sensitive assays for antigen-induced soluble fac-tors and the proper tools for biochemical studies resolving theirnature and mechanism of action.

We thank H. Skarvall, R. Tees, and B. Geschke for technical assis-tance; L. Forni for immunofluorescent reagents; W. Raschke forW279. 1 cells; J. Albertini, W. Breisinger, and C. Nordstrom for pre-paring the manuscript; and K. Smith and F. Melchers for critical readingand comments. The Basel Institute for Immunology was founded by andis supported by F. Hoffmann-La Roche and Co., CH-4002 Basel,Switzerland.

1. Schimpl, A. & Wecker, E. (1979) in Biology ofthe Lymphokines,eds. Cohen, S., Pick, E. & Oppenheim, J. J. (Academic, NewYork), pp. 369-390.

2. Feldman, M., Howie, S. & Kontiainen, S. (1979) in Biology ofthe Lymphokines, eds. Cohen, S., Pick, E. & Oppenheim, J. J.(Academic, New York), pp. 391-419.

3. Schreier, M. H. (1981) in Lymphokines, ed. Pick, E. (Academic,New York), Vol. 2, pp. 31-61.

4. Schreier, M. H., Andersson, J., Lernhardt, W. & Melchers, F.(1980)J. Exp. Med. 151, 194-203.

5. Melchers, F., Andersson, J., Lernhardt, W. & Schreier, M. H.(1980) Eur. J. Immunol 10, 679-685.

6. Raff, M. C., Megson, M., Owen, J. J. T. & Cooper, M. D. (1976)Nature (London) 259, 224-226.

7. Burrows, M., Lejeune, M. & Kearney, J. F. (1979) Nature (Lon-don) 280, 838-841.

8. Maki, R., Kearney, J., Paige, C. & Tonegawa, S. (1980) Science209, 1366-1369.

9. Perry, R. P., Kelly, D. E., Coleclough, C. & Kearney, J. F.(1981) Proc. Nati Acad. Sci. USA 78, 247-251.

10. Paige, C. J., Kincade, P. W. & Ralph, P. (1981) Nature (London)292, 631-633.

11. Gordon, J., Hamblin, T. J., Smith, J. C., Stevenson, F. K. &Stevenson, G. T. (1981) Blood 58, 552-556.

12. Osmond, D. G. & Nossal, G. J. V. (1974) Cell Immunol. 13,132-145.

13. Sidman, C. (1981) Cell 23, 379-389.14. Alt, F. W., Bothwell, A. L., Knapp, M., Siden, E., Mather, E.,

Koshland, M. & Baltimore, D. (1980) Cell 20, 293-301.15. Early, P., Rogers, F., Davis, M., Calame, K., Bond, M., Wall,

R. & Hood, L. (1980) Cell 20, 313-319.16. Paige, C. J., Kincade, P. W. & Ralph, P. (1978)J. Immunol. 121,

641-647.17. Harris, A. W. (1977) in Protides of the Biological Fluids, ed. Pe-

ters, H. (Pergamon, Oxford, England), Vol. 25, pp. 601-604.18. Sibley, C. H., Ewald, S. J, Kehry, M. R., Douglas, R. H.,

Raschke, W. C. & Hood, L. E. (1980) J. Immunol. 125,2097-2105.

19. Baines, P., Dexter, T. M. & Schofield, R. (1979) Leuk. Res. 3,23-29.

20. Warner, N. L., Harris, A. W. & Gutman, G. A. (1975) in Mem-brane Receptors of Lymphocyte, eds. Seligmann, M.,Preud'homme, J. L. & Kourilsky, F. M. (North-Holland, Am-sterdam), pp. 203-216.

21. Raschke, W. C. (1978) Curr. Top. Microbiol Immunol 81, 70-76.22. Bergman, Y. & Haimovich, J. (1977) Eur. J. Immunol 7, 413-

417.23. Schreier, M. H. & Tees, R. (1980) Int. Arch. Allergy Appl. Im-

munol 61, 227-237.24. Schreier, M. H., Tees, R. & Nordin, A. A. (1982) in Lympho-

kines, eds. Feldmann, M. & Schreier, M. H. (Academic, NewYork), Vol. 5, pp. 443-464.

25. Gronowicz, E., Coutinho, A. & Melchers, F. (1976) Eur. J. Im-munol. 6, 588-590.

26. Schreier, M. H., Iscove, N. N., Tees, R., Aarden, L. & von

Boehmer, H. (1980) Immunol Rev. 51, 315-336.27. Nabel, G., Greenberger, J. S., Sakakeeny, M. A. & Cantor, H.

(1981) Proc. Natl. Acad. Sci. USA 78, 1157-1161.28. Schrader, J. W., Arnold, B. & Clark-Lewis, I. (1980) Nature

(London) 283, 197-199.29. Schreier, M. & Iscove, N. N. (1980) Nature (London) 287,

228-230.30. Perry, R. P. & Kelley, D. E. (1979) Cell 18, 1333-1339.31. Sakaguchi, N., Kishimoto, T., Kikutani, H., Watanabe, T., Yo-

shida, N., Shimizu, A., Yamawaki-Kataoka, Y., Honjo, T. & Ya-mamura, Y. (1980) J. Immunol. 125, 2654-2659.

32. Maino, V. C., Kurnick, J. T., Kubo, R. T. & Grey, H. M. (1977)J. Immunol. 118, 742-748.

33. Fu, S. M., Chiorazzi, N., Kunkel, H. G., Halper, J. P. & Harris,S. R. (1978)J. Exp. Med. 148, 1570-1578.

34. Fu, S. M., Chiorazzi, N. & Kunkel, H. G. (1979) Immunol. Rev.48, 23-44.

35. Siden, F. L., Baltimore, D., Clark, D. & Rosenberg, N. E.(1979) Cell 16, 389-396.

36. Boss, M., Greaves, M. & Teich, N. (1979) Nature (London) 278,551-553.

37. Knapp, M. R., Gronowicz, E. S., Schroder, J. & Strober, S.(1979)J. Immunol 123, 1000-1006.

38. Strober, S., Gronowicz, E. S., Knapp, M. R., Slavin, S., Vitetta,E. S., Warnke, R. A., Kotzin, B. & Schroder, J. (1979) ImmunolRev. 48, 169-195.

39. Boyd, A. W., Goding, J. W. & Schrader, J. W. (1981)J. Immunol126, 2461-2465.

40. Saiki, O., Kishimoto, T., Muraguchi, A. M. & Yamamura, Y.(1980)J. Immunol 124, 2609-2614.

41. Schimpl, A., Hebner, L., Wong, C. A. & Wecker, E. (1980)Behring Inst. Mitt. 67, 221-225.

42. Muraguchi, A., Kishimoto, T., Mild, Y., Kuritani, T., Kaieda, T.,Yoshizaki, K. & Yamamura, Y. (1981)J. Immunot 127, 412-416.

4760 Immunology: Paige et aL

Dow

nloa

ded

by g

uest

on

Aug

ust 1

5, 2

021